cross-lagged networks
Definition
Cross-lagged networks refers to a longitudinal panel modeling approach that examines direct predictive relationships between individual observed variables (e.g., symptoms, attitudes, or scale items) over discrete measurement occasions, rather than between latent constructs. The method combines regularized regression estimation with structural equation modeling to estimate auto-regressive and cross-lagged pathways that characterize how individual components of psychological constructs influence each other over time. By analyzing item-level longitudinal effects within and across constructs, cross-lagged networks can reveal specific pathways that are often obscured by traditional latent variable approaches, thereby generating more detailed and testable causal hypotheses. This approach is particularly well-suited for panel data collected at a small set of measurement occasions from large samples, especially when relationships are non-stationary or when empirical evidence suggests that direct item-level relations better explain covariance than underlying latent variables.
Sources: Wysocki et al. (2025)
Related Terms
Applications
Cross-lagged Networks and Developmental Research
Cross-lagged networks are particularly valuable for developmental research, which uses panel designs to investigate how variables relate to each other over developmental periods. The method allows researchers to model longitudinal associations among individual elements of constructs at discrete measurement occasions separated by months or years.
Sources: Wysocki et al. (2025)
Cross-lagged Networks and Psychological Constructs
Cross-lagged networks extend network modeling approaches to longitudinal panel data, enabling investigation of how individual items measuring psychological constructs predict one another over time. This item-level approach reveals direct influences between observed components rather than modeling associations through latent variables.
Sources: Wysocki et al. (2025)
